feat: update the readme
Browse files
README.md
CHANGED
|
@@ -1,3 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
Core implementation of Jina XLM-RoBERTa
|
| 2 |
|
| 3 |
This implementation is adapted from [XLM-Roberta](https://huggingface.co/docs/transformers/en/model_doc/xlm-roberta). In contrast to the original implementation, this model uses Rotary positional encodings and supports flash-attention 2.
|
|
@@ -9,4 +111,4 @@ to be added soon
|
|
| 9 |
|
| 10 |
### Converting weights
|
| 11 |
|
| 12 |
-
Weights from an [original XLMRoberta model](https://huggingface.co/FacebookAI/xlm-roberta-large) can be converted using the `convert_roberta_weights_to_flash.py` script in the model repository.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- transformers
|
| 4 |
+
- xlm-roberta
|
| 5 |
+
library_name: transformers
|
| 6 |
+
license: cc-by-nc-4.0
|
| 7 |
+
language:
|
| 8 |
+
- multilingual
|
| 9 |
+
- af
|
| 10 |
+
- am
|
| 11 |
+
- ar
|
| 12 |
+
- as
|
| 13 |
+
- az
|
| 14 |
+
- be
|
| 15 |
+
- bg
|
| 16 |
+
- bn
|
| 17 |
+
- br
|
| 18 |
+
- bs
|
| 19 |
+
- ca
|
| 20 |
+
- cs
|
| 21 |
+
- cy
|
| 22 |
+
- da
|
| 23 |
+
- de
|
| 24 |
+
- el
|
| 25 |
+
- en
|
| 26 |
+
- eo
|
| 27 |
+
- es
|
| 28 |
+
- et
|
| 29 |
+
- eu
|
| 30 |
+
- fa
|
| 31 |
+
- fi
|
| 32 |
+
- fr
|
| 33 |
+
- fy
|
| 34 |
+
- ga
|
| 35 |
+
- gd
|
| 36 |
+
- gl
|
| 37 |
+
- gu
|
| 38 |
+
- ha
|
| 39 |
+
- he
|
| 40 |
+
- hi
|
| 41 |
+
- hr
|
| 42 |
+
- hu
|
| 43 |
+
- hy
|
| 44 |
+
- id
|
| 45 |
+
- is
|
| 46 |
+
- it
|
| 47 |
+
- ja
|
| 48 |
+
- jv
|
| 49 |
+
- ka
|
| 50 |
+
- kk
|
| 51 |
+
- km
|
| 52 |
+
- kn
|
| 53 |
+
- ko
|
| 54 |
+
- ku
|
| 55 |
+
- ky
|
| 56 |
+
- la
|
| 57 |
+
- lo
|
| 58 |
+
- lt
|
| 59 |
+
- lv
|
| 60 |
+
- mg
|
| 61 |
+
- mk
|
| 62 |
+
- ml
|
| 63 |
+
- mn
|
| 64 |
+
- mr
|
| 65 |
+
- ms
|
| 66 |
+
- my
|
| 67 |
+
- ne
|
| 68 |
+
- nl
|
| 69 |
+
- 'no'
|
| 70 |
+
- om
|
| 71 |
+
- or
|
| 72 |
+
- pa
|
| 73 |
+
- pl
|
| 74 |
+
- ps
|
| 75 |
+
- pt
|
| 76 |
+
- ro
|
| 77 |
+
- ru
|
| 78 |
+
- sa
|
| 79 |
+
- sd
|
| 80 |
+
- si
|
| 81 |
+
- sk
|
| 82 |
+
- sl
|
| 83 |
+
- so
|
| 84 |
+
- sq
|
| 85 |
+
- sr
|
| 86 |
+
- su
|
| 87 |
+
- sv
|
| 88 |
+
- sw
|
| 89 |
+
- ta
|
| 90 |
+
- te
|
| 91 |
+
- th
|
| 92 |
+
- tl
|
| 93 |
+
- tr
|
| 94 |
+
- ug
|
| 95 |
+
- uk
|
| 96 |
+
- ur
|
| 97 |
+
- uz
|
| 98 |
+
- vi
|
| 99 |
+
- xh
|
| 100 |
+
- yi
|
| 101 |
+
- zh
|
| 102 |
+
---
|
| 103 |
Core implementation of Jina XLM-RoBERTa
|
| 104 |
|
| 105 |
This implementation is adapted from [XLM-Roberta](https://huggingface.co/docs/transformers/en/model_doc/xlm-roberta). In contrast to the original implementation, this model uses Rotary positional encodings and supports flash-attention 2.
|
|
|
|
| 111 |
|
| 112 |
### Converting weights
|
| 113 |
|
| 114 |
+
Weights from an [original XLMRoberta model](https://huggingface.co/FacebookAI/xlm-roberta-large) can be converted using the `convert_roberta_weights_to_flash.py` script in the model repository.
|